A zirconium-mediated synthesis of (±)-tecomanine

Article abstract of DOI:10.1055/s-1998-5924

Intramolecular cocyclisation of N-(but-2-enyl)-N-methyl-N-(2-methylbut- 3-ynyl)amine using zirconocene(ethylene), followed by carbonylation affords a 4 : 1 mixture of (±)-tecomanine and (±)-4-epi-tecomanine.

Full text of DOI:10.1055/s-1998-5924

SYNTHESIS
Papers
552
A Zirconium-Mediated Synthesis of (±)-Tecomanine
Mark I. Kemp,a Richard J. Whitby,*a Steven J. Cooteb
a
Department of Chemistry, The University, Southampton, Hants, SO17 1BJ, U.K.
Fax +44(1703)593781; E-mail: rjw1@soton.ac.uk
Glaxo Wellcome Medicines Research Centre, Gunnels Wood Road, Stevenage, Herts, SG1 2NY, U.K.
b
Received 22 August 1997
Abstract: Intramolecular cocyclisation of N-(but-2-enyl)-N-
methyl-N-(2-methylbut-3-ynyl)amine using zirconocene(ethyl-
ene), followed by carbonylation affords a 4 : 1 mixture of (±)-
tecomanine and (±)-4-epi-tecomanine.
Key words: tecomanine, 4-epi-tecomanine, zirconium-induced
cocyclizations of enynes
Figure. Transition States Leading to 2/epi-2 (R = H)
Tecomanine [(–)-1], the major alkaloid of Tecoma stans
(Juss.), was isolated by Hammouda and Motawi1 in 1959.
The structure of the monoterpene-based alkaloid was elu-
cidated in 1963 by Jones et al.2 whilst its absolute stereo-
chemistry was found by X-ray crystal structure analysis in
1971.3 Hammouda et al.4 have shown that tecomanine
possesses potent hypoglycaemic activity, confirming the
reputation of the plant leaves as an antidiabetic drug by
the Mexican people.5 There have been two previous syn-
theses of (±)-tecomanine, and one of (+)-tecomanine.6
Our success in the zirconocene-induced cocyclisation of
4-aza-1,7-enynes to form 3,4-disubstituted piperidines on
protonolysis7 suggested that tecomanine could be derived
concisely from the enyne 3 by zirconocene induced cocy-
clisation to give 2 followed by carbonylation.8-10
and the zirconocene(alkyne) complex so-formed may be
trapped by other species. Based on this precedent, we re-
acted the model enyne 5 with zirconocene(ethylene) (4),
generated in situ from diethylzirconocene,12 to give the
zirconacyclopentenes 6 and 7 (~ 3 : 1, Scheme 1). Heating
6 + 7 to reflux in tetrahydrofuran eliminated ethylene, pre-
sumably to form the zirconacyclopropene 8 which under-
went intramolecular carbometallation to give the desired
zirconabicycle 9. After 5 hours, aqueous workup gave the
desired piperidine 10 and the diene 11 in 35 and 14%
yield, respectively. The lack of a product derived from 7
suggests that it loses ethylene faster than 6. The reaction
was repeated with heating to reflux for 24 hours to give 10
as the sole product in a 50% yield on aqueous workup.
The successful formation of 10 proved for the first time
that zirconocene(ethylene) was able to overcome one of
the major limitations of zirconocene(but-1-ene) in in-
tramolecular couplings. Since completing this work the
successful cyclisation of enynes containing a terminal
alkyne moiety using Cp2ZrCl2/Mg/Hg has been de-
scribed,13 as has the successful cyclisation of N-allyl-N-
propargylbenzylamine using (i-PrO)2Ti(η2-propene).14
An immediate problem with the proposed route was the
presence of a terminal alkyne in 3 since these are known
to fail to cyclise with the usual zirconocene equivalent
[zirconocene(but-1-ene) - the Negishi reagent].9 The usual
solution to this limitation is to protect the terminal alkyne
as the trimethylsilyl or trimethylstannyl derivative.9 Un-
fortunately, consideration of the transition states for cycli-
sation (Figure) suggested that with a bulky group on the
alkyne terminus (R = SiMe3 or SnMe3) transition state A
is destabilised relative to B by a severe 1,3-allylic interac-
tion leading to the wrong stereochemistry of cyclisation
for tecomanine.11 In the cyclisation of 3 1,3-diaxial inter-
actions in the transition state B where the methyl group is
axial destabilises it relative to the transition state A where
the 1,3-allylic interaction is now negligible (R = H) lead-
ing to the correct stereochemistry for tecomanine.
Scheme 1, R = (CH2)2CH=CH2
We therefore required a source of ‘Cp2Zr’ that would tol-
erate a terminal alkyne. Takahashi et al.12 have shown that
zirconocene(ethylene) (4) inserts a range of alkynes in-
cluding oct-1-yne – a terminal alkyne. On heating the re-
sultant zirconacyclopentenes to 50°C ethylene is extruded
We were now ready to attempt the total synthesis of (±)-
tecomanine. The required cyclisation precursor, enyne 3,
was synthesised as summarised in Scheme 2. N-Methyltri-
fluoroacetamide (12) was alkylated with commercial 1-

SYNTHESIS
April 1998
553
chlorobut-2-ene (85% E) followed by basic hydrolysis to
give N-(but-2-enyl)-N-methylamine (14).15 (±)-But-3-yn-
2-ol was converted to the tosylate 1516 which was heated
with lithium bromide in dimethyl sulphoxide at 80°C and
20 mbar to give the bromide 16 as the distillate in good
yield. Reaction of 16 with aluminum (0.66 equiv) and a
trace of mercury(II) chloride gave the mixed allenic orga-
noaluminum 17, which was reacted with paraformalde-
hyde to yield the homologated alcohol 18.17 The alcohol
18 could not be isolated pure since it formed an azeotrope
with tetrahydrofuran so it was converted to the tosylate
19, obtained in 65% overall yield from 16. Reaction be-
tween N-(but-2-enyl)-N-methylamine (14) and the tosy-
late 19 gave the required enyne 3, although 2.1
equivalents of the amine was required for good conver-
sion despite the presence of excess potassium carbonate.
Scheme 3
4-epi-tecomanine (26)18 in a disappointing 21% overall
yield from 3. The yield was slightly increased to 31% by
carrying out the carbonylation at –78°C for 2 hours then
working up with iodine (2 equiv). Of note is that products
from the (Z)-but-2-enyl moiety were not isolated. Epimer-
isation during carbonylation of the supposed intermediate
zirconocene–η2-ketone complex 27 via the hydride 2819 is
most likely.
Scheme 2
Reaction of the enyne 3 with in situ generated zir-
conocene(ethylene) (4) at room temperature formed the
monocyclic complexes 20 and 21 (3 : 1) (Scheme 3).
Heating at reflux in tetrahydrofuran for 3 hours gave com-
plete conversion to the zirconabicycles 22 and 23 and
hence the piperidines 24 and 25 (4 : 1, 46% yield) on
methanolysis. The major piperidine 24 was easily as-
signed as cis-disubstituted as judged from coupling pat-
terns of the four methylene hydrogens adjacent to nitrogen
The identities of 1 and 26 were confirmed by comparison
of their 1H, 13C, IR, UV, and MS data with literature val-
ues.6a,18 Furthermore, the picrate of tecomanine was
formed and its melting point (184–186°C) found to be in
good agreement with that reported6a (184.5–185.5°C).
We also confirmed that epi-7,7a-tecomanine (26) could be
epimerised to tecomanine (1) in ~50% yield by stirring a
benzene solution over basic alumina overnight as suggest-
ed by Miyashita et al.6b
1
in the H NMR spectra. H-2ax and H-6ax were triplets
(J = 11 Hz) and H-2eq and H-6eq were dd (J = 11, 4 Hz),
showing that both the methyl and ethyl groups must be
equatorial. The predominant formation of 24 confirmed
that the major zirconabicycle was 22, as required for the
synthesis of tecomanine. It should be noted that although
the room temperature zirconocene(but-1-ene)-induced
cocyclisation of enynes gives the kinetic product, the
same may not be true of the high temperature zir-
conocene(ethylene) route used to cyclise 3. Consideration
of the relative stabilities of the zirconacycle products de-
rived from the cyclisations shown in the Figure indicates
that the thermodynamic product will be the same as the ki-
netic, and for the same reasons.
The low yield of the zirconocene-induced cocyclisation /
carbonylation step in our synthesis of tecomanine has two
causes. The zirconocene(ethylene)-induced cocyclisation,
which requires heating, gives lower yields of zirconacy-
cles than zirconocene(but-1-ene). It is also known that
carbonylation of unsaturated zirconacycles only gives
high yields with bulky vinyl substituents.9 Both problems
would be overcome by using a trimethylsilyl or trimethyl-
stannyl group to protect the terminal alkyne, so it was
worth checking that the unfavourable prediction for stereo-
control in the cyclisation (Figure) was true. A suitable
enyne 29 was easily prepared and cocyclised with zir-
Carbonylation (1 atm CO, r.t., 16 h) of 22/23 gave an eas-
ily separated 4 : 1 mixture of (±)-tecomanine (1) and (±)-

SYNTHESIS
Papers
554
EI-MS: m/z = 201 (M+, 100%), 200 (70), 186 (69), 159 (19), 124 (18),
110 (30), 91 (70).
conocene(but-1-ene) to give a single zirconacycle 30.
Protonolysis gave exclusively the trans-substituted pi-
peridine 31 showing that, as expected, the zirconacycle
formed had the wrong relative sterochemistry for tecoma-
nine. The trans stereochemistry of 31 followed from cou-
pling patterns of the methylene hydrogens adjacent to
nitrogen (see experimental).
HRMS (EI): m/z found 201.1521 (M+), C14H19N requires 201.1517.
11:
IR (film): ν = 3076 m, 3027 m, 2964 s, 2934 s, 2796 s, 1641 m, 1602
w, 1494 m, 1453 s, 1368 m, 899 s, 738 s, 698 s cm–1.
1H NMR (270 MHz, CDCl3): δ = 7.20–7.30 (5 H, m), 5.76 (1 H, ddt,
J = 17, 10, 7 Hz), 5.03 (1 H, d, J = 17 Hz), 4.96 (1 H, d, J = 10 Hz),
4.95 (1 H, s), 4.84 (1 H, s), 3.52 (2 H, s), 2.96 (2 H, s), 2.44 (2 H, t, J
= 7 Hz), 2.23 (2 H, q, J = 7 Hz), 2.09 (2 H, q, J = 7 Hz), 0.98 (3 H, t,
J = 7 Hz).
13C NMR (67.5 MHz, CDCl3): δ = 149.84 (0), 140.31 (0), 137.36 (1),
128.89 (1), 128.26 (1), 126.83 (1), 115.40 (2), 110.76 (2), 60.02 (2),
58.38 (2), 53.16 (2), 31.70 (2), 26.96 (2), 12.34 (3).
EI-MS: m/z (%) = 214 (M–15+, 4), 188 (M–41+ [loss of allyl], 83), 91
(100), 41 (17).
Toluene-4-sulphonic Acid 2-Methylbut-3-ynyl Ester (19):
A THF solution of 1817 (3.1 g, 36 mmol) was added to a solution of
toluene-4-sulphonyl chloride (8.4 g, 44 mmol) in Et2O (50 mL).
Freshly ground KOH (20 g) was added in small portions over 30 min
whilst maintaining the temperature <0°C. The heterogenous mixture
was then stirred at 0°C for 30 min before pouring onto iced water
(50 mL). The layers were seperated and the aqueous layer extracted
with Et2O (2 × 20 mL). The combined organic layers were dried
(MgSO4), solvent removed, and the residue chromatographed (petrol/
Et2O, 95:5) to afford 19 as a clear colourless oil (8.3 g, 35 mmol, 97%).
IR (film): ν = 3290 s, 2983 m, 1599 m, 1496 w, 1457 m, 1360 s, 1177
s, 978 s, 665 s cm–1.
In conclusion, we have shown how the zirconocene-in-
duced cocyclisation of an enyne followed by carbonyla-
tion provides a short synthesis of the hypoglycaemic
alkaloid (±)-tecomanine. A method allowing the zir-
conocene-induced cocyclisation of terminal alkynes has
also been developed.
For general experimental details, and the preparation of 5, see previ-
ous paper.7b 2-Methylbut-3-yn-1-ol (18) was prepared by the litera-
ture procedure.17
1H NMR (270 MHz, CDCl3): δ = 7.81 (2 H, d, J = 8 Hz), 7.35 (2 H,
d, J = 8 Hz), 4.03 (1 H, dd, J = 9, 7 Hz), 3.88 (1 H, dd, J = 9, 7 Hz),
2.78 (1 H, sextet of d, J = 7, 2 Hz), 2.46 (3 H, s), 2.05 (1 H, d, J =
2 Hz), 1.19 (3 H, d, J = 7 Hz).
1-Benzyl-4-methyl-3-methylenepiperidine (10) and N-Benzyl-N-
(but-3-enyl)-N-(2-ethylallyl)amine (11):
13C NMR (75 MHz, CDCl3): δ = 145.10 (0), 132.91 (0), 130.01 (1),
128.08 (1), 83.53 (1), 72.48 (2), 70.65 (0), 26.13 (1), 21.76 (3), 17.22
(3).
EtMgCl (2 M solution in THF, 0.5 mL, 1 mmol) was added to
Cp2ZrCl2 (146 mg, 0.5 mmol) in THF (2 mL) at –78°C under argon.
After 1 h at –78°C, N-benzyl-N-(but-3-enyl)-N-(prop-2-ynyl)amine7b
(5; 100 mg, 0.5 mmol) in THF (1 mL) was added and the yellow so-
lution warmed to r.t. As the mixture reached 0°C it became deep red.
After 2 h at r.t., the solvent was removed in vacuo and benzene (10
mL) was added to the residue. The resultant cloudy benzene solution
was filtered under argon and concentrated in vacuo. C6D6 (0.5 mL)
was added and the solution transferred under argon to an NMR tube.
The tube was suspended above EtOH heated to reflux (78°C) and
analysed after 5 h. Interpretation of the 1H NMR spectrum was ham-
pered by the presence of minor products, but the 13C spectrum clearly
showed 9 to be the major zirconacycle.
EI-MS: m/z (%) = 238 (M+, 5), 155 (100), 91 (86), 65 (27).
N-[(E)- and (Z)-But-2-enyl]-2,2,2-trifluoro-N-methylacetamide
(13):
A solution of 12 (900 mg, 7.08 mmol) in THF (5 mL) was added to a
suspension of KH (340 mg, 8.50 mmol) in THF (10 mL) at 0°C. The
mixture was stirred until effervescence subsided (~ 5 min). 18-
Crown-6 (15 mg, 0.057 mmol) and 1-chlorobut-2-ene (85% E,
900 mg, 10 mmol) was added and the mixture heated to reflux for
16 h. The mixture was cooled and poured onto 1% HCl (15 mL), Et2O
(50 mL) was added and the layers separated. The aqueous layer was
extracted with Et2O (2 × 20 mL) and the combined organic layers
washed with H2O (2 × 20 mL), dried (MgSO4) and solvent removed
to give a yellow oil. The crude product was purified using column
chromatography (petrol/Et2O, 95:5) and Kugelrohr distillation (oven
temp ~ 70°C/10 Torr) to afford 13 (85% E) as a clear colourless oil
(794 mg, 4.39 mmol, 62%).
13C NMR (75 MHz, C6D6): δ = 172.21 (1), 149.79 (0), 140.52 (0),
130.07 (1), 129.09 (1), 127.68 (1), 112.75 (1), 111.87 (1), 65.95 (2),
64.57 (2), 55.18 (2), 48.40 (2), 42.17 (1), 40.21 (2).
MeOH (1 mL) was added to the NMR tube and the resultant pale-yel-
low solution was poured onto sat. NaHCO3 solution (5 mL) and Et2O
(10 mL). Extraction into Et2O, drying (K2CO3), and removal of sol-
vent afforded a cloudy yellow oil. Column chromatography (grade II
neutral alumina, petrol/Et2O, 95:5) allowed the isolation of 10 (35
mg, 0.18 mmol, 35%) and 11 (15 mg, 0.07 mmol, 14%). When the re-
action was repeated, but heated to reflux in THF for 24 h before
quenching, 10 was the only product in 50% NMR yield (hexamethyl-
benzene as standard).
IR (film): ν = 3014 w, 2944 m, 2860 w, 1697 s, 1517 w, 1490 m,
1452 m, 1422 m, 1380 w, 1358 w, 1194 s, 1144 s, 967 s cm–1.
1H NMR (270 MHz, CDCl3), complicated by the presence of two ro-
tomers for each of the two isomers: δ = 5.70 (1 H, m), 5.40 (1 H, m),
4.11, 4.09 (0.3 H combined, two doublets, J = 9 Hz), 3.95, 3.97 (1.7
H combined, two doublets, J = 6 Hz), 3.09, 2.97 (3 H combined, two
singlets), 1.74, 1.72 (3 H combined, two singlets).
10:
13C NMR (67.5 MHz, CDCl3), complicated by the presence of two ro-
tomers for each of the two isomers: δ = 156.68 (0, q, J = 36 Hz due to
coupling with 19F), 131.30 (1), 131.04 (1), 130.07 (1), 129.68 (1),
124.41 (1), 123.87 (1), 123.59 (1), 123.02 (1), 116.65 (0, q, J = 284
Hz due to coupling with 19F), 51.43 (2), 50.85 (2), 45.90 (2), 45.21
(2), 33.92 (3), 33.62 (3), 17.62 (3), 12.87 (3).
IR (film): ν = 3026 w, 2958 m, 2931 s, 2872 m, 2846 m, 2793 m, 2747
m, 1652 m, 1599 w, 1495 m, 1453 m, 1407 w, 1365 m, 1336 m, 1306
m, 698 s cm–1.
1H NMR (300 MHz, CDCl3): δ = 7.20–7.30 (5 H, m), 4.77 (1 H, s),
4.72 (1 H, s), 3.52 (2 H, s), 3.26 (1 H, d, J = 12 Hz), 2.87 (1 H, d + fs,
J = 11 Hz), 2.63 (1 H, d, J = 12 Hz), 2.19 (1 H, td, J = 11, 4 Hz), 2.07
(1 H, m), 1.70 (1 H, dq, J = 11, 5 Hz), 1.34 (1 H, qd, J = 11, 4 Hz),
1.08 (3 H, d, J = 7 Hz).
CI-MS (NH3): m/z = 199 (MNH4+), 182 (MH+).
13C NMR (75 MHz, CDCl3): δ = 149.21 (0), 138.41 (0), 129.42 (1),
128.32 (1), 127.12 (1), 107.19 (2), 63.03 (2), 61.00 (2), 53.51 (2),
35.73 (1), 34.76 (2), 17.89 (3).
N-(But-2-enyl)-N-methylamine (14):
Compound 13 (10 g, 55 mmol) was added to a suspension of K2CO3
in 80% aq MeOH (60 mL) and stirred overnight at r.t. HCl gas was

SYNTHESIS
April 1998
555
bubbled through the mixture for 30 min, then the solution concentrat-
ed to give a white residue which was washed with Et2O (3 × 50 mL).
H2O (5 mL) was added followed by NaOH until the pH was 12. Sep-
aration of the resultant top layer gave 14 (4.21 g, 49.5 mmol, 90%,
85% E). The 1H NMR and IR data are in close agreement with that in
the literature.20
(4R*,7S*,7aS*)-2,3,4,6,7,7a-Hexahydro-2,4,7-trimethyl-1H-
cyclopenta[c]pyridin-6-one [(±)-Tecomanine] (1) and
(4R*,7S*,7aR*)-2,3,4,6,7,7a-Hexahydro-2,4,7-trimethyl-1H-
cyclopenta[c]pyridin-6-one [(±)-epi-7,7a-Tecomanine] (26):
The zirconabicycle 2 was prepared from 3 (151 mg, 1 mmol) by the
same procedure used in the formation of 24. After 3 h at reflux, the
solution was cooled to –78°C and the Schlenk flask was evacuated
and refilled with CO under a small positive pressure using a double-
walled balloon. After stirring at –78°C for 2 h, I2 (508 mg, 2 mmol)
in THF (2 mL) was added and the reaction was warmed to r.t. MeOH
(1 mL) was added and the CO atmosphere was removed. Sat.
NaHCO3 solution (5 mL) and Et O (20 mL) were added and the mix-
ture was thoroughly shaken. Th2e two layers were separated and the
aqueous phase was extracted with Et2O (3 × 20 mL). The combined
organic layers were washed with sat. Na2S2O3 soln (2 × 10 mL), dried
(K2CO3), filtered and concentrated in vacuo to afford a dark-brown
oil. The crude product was purified using column chromatography
(grade II neutral alumina, hexane/Et2O, 50:50) to afford 26 (10 mg,
0.06 mmol, 6%) and 1 (45 mg, 0.25 mmol, 25%) as clear, colourless
viscous oils.
IR (film): ν = 3314 br, 2919 s, 2856 m, 2796 m, 1654 m, 1541 w, 1451
m, 1378 m, 1353 w, 969 s cm–1.
1H NMR (270 MHz, CDCl3): δ = 5.40–5.60 (2 H, m), 3.23 (0.3 H, d,
J = 7 Hz), 3.12 (1.7 H, d, J = 6 Hz), 2.42 (0.45 H, s), 2.40 (2.55 H, s),
2.40 (2.55 H, d, J = 6 Hz), 1.66 (0.45 H, partially obscured doublet).
13C NMR (67.5 MHz, CDCl3): δ = (Z): 128.63 (1), 126.40 (1), 47.96
(2), 35.99 (3), 13.09 (3). (E): 129.36 (1), 127.54 (1), 53.78 (2), 35.83
(3), 17.86 (3).
EI-MS: m/z (%) = 85 (M+, 39), 84 (36), 70 (100), 55 (28), 44 (94), 30
(36).
N-[(E)- and (Z)-But-2-enyl]-N-methyl-N-(2-methylbut-3-
ynyl)amine (3):
A solution of 19 (2.38 g, 10 mmol) in MeCN (10 mL) was added to
14 (1.79 g, 21 mmol) and K2CO3 (1.6 g, 12 mmol) in MeCN (10 mL)
and the mixture was heated to reflux for 24 h. After cooling, H2O
(40 mL) and Et2O (40 mL) were added, the layers separated, and the
aqueous layer extracted with Et2O (2 × 20 mL). The combined organ-
ic layers were dried (K2CO3) and solvent removed to yield a yellow
oil. The crude product was purified using column chromatography
(grade II neutral alumina, petrol/Et2O, 96:4) and Kugelrohr distilla-
tion (60°C/10 Torr) to afford 3 as a clear colourless oil (695 mg,
4.6 mmol, 46%).
1 (data are in good agreement with that reported6):
IR (film): ν = 2962 m, 2934 m, 2876 m, 2781 m, 1704 s, 1620 m, 1464
m, 1373 m cm–1.
UV (EtOH): λmax = 224 nm (log ε = 3.98).
1H NMR (270 MHz, CDCl3): δ = 5.85 (1 H, s, H-5), 3.17 (1 H, ddd,
J = 11, 6, 2 Hz, H-1eq), 2.94 (1 H, ddd, J = 11, 6, 2 Hz, H-3eq), 2.64
(1 H, d pentet + fs, J = 11.2, 6 Hz, H-4), 2.50 (1 H, ddddd, J = 11.4,
6, 3, 1.7, 1 Hz, H-7a), 2.27 (3 H, s, N-CH3), 1.89 (1 H, qd, J = 8, 3 Hz,
H-7), 1.74 (2 H, t, J = 11 Hz, H-1ax and H-3ax), 1.19 (3 H, d, J = 7
Hz, 7-CH3), 1.16 (3 H, d, J = 6 Hz, 4-CH3).
IR (film): ν = 3309 m, 2970 s, 2934 s, 2841 m, 2793 s, 2113 w, 1670
w, 1457 m, 1377 m, 1336 w, 967 s, 630 s cm–1.
1H NMR (270 MHz, CDCl3): δ = 5.40–5.70 (2 H, m), 3.06 (0.3 H, d,
J = 7 Hz), 2.97 (1.7 H, d, J = 5 Hz), 2.61 (1 H, sextet of d, J = 7, 2 Hz),
2.47 (1 H, dd, J = 12, 7 Hz), 2.27 (1 H, obscured), 2.26 (0.45 H, s),
2.24 (2.55 H, s), 2.07 (1 H, d, J = 2 Hz), 1.69 (2.55 H, d, J = 6 Hz),
1.64 (0.45 H, d, J = 7 Hz), 1.18 (3 H, d, J = 7 Hz).
13C NMR (75 MHz, CDCl3): δ = 210.63 (0), 183.68 (0), 124.39 (1),
63.25 (2), 62.05 (2), 49.70 (1), 45.63 (3), 45.16 (1), 35.24 (1), 15.06
(3), 15.00 (3).
EI-MS: m/z (%) = 180 (15), 179 (M+, 100), 164 (32), 111 (35), 93
(43), 58 (39), 57 (94), 42 (46).
13C NMR (75 MHz, CDCl3): δ = (Z): 127.51 (1), 126.99 (1), 88.42
(1), 68.64 (0), 62.99 (2), 54.30 (2), 42.82 (3), 24.70 (1), 19.20 (3),
13.25 (3); (E): 128.71 (1), 128.27 (1), 88.42 (1), 68.64 (0), 62.78 (2),
60.42 (2), 42.65 (3), 24.70 (1), 19.20 (3), 17.93 (3).
HRMS (EI): m/z found 179.1306 (M+), C11H17NO requires 179.1310.
The picrate was formed by adding picric acid (0.9 equiv) in EtOH to
1 in EtOH. Recrystallisation (EtOH) afforded the picrate as yellow
plates; mp 184–186°C (Lit.6a 184.5–185.5°C).
CI-MS (NH3): m/z = 152 (MH+).
Analysis: found C, 50.29; H, 4.76; N, 13.52; C11H17NO·C6H3N3O7
HRMS (EI): m/z found 151.1375 (M+), C10H17N requires 151.1361.
requires C, 50.00; H, 4.90; N, 13.73.
(3S*,5R*)-3-Ethyl-1,5-dimethyl-4-methylenepiperidine (24):
EtMgBr (3 M soln in Et2O, 0.46 mL, 1.39 mmol) was added to
Cp2ZrCl2 (213 mg, 0.73 mmol) in THF (2 mL) at –78°C under argon.
After 1 h at –78°C, 3 (100 mg, 0.66 mmol) in THF (2 mL) was added
and the mixture was warmed to r.t. The resultant orange-red solution
was heated to reflux for 3 h and became dark brown. After cooling,
MeOH (1 mL) and sat. NaHCO3 solution (5 mL) were added, fol-
lowed by Et2O (10 mL). The organic layer was combined with further
extractions (Et2O, 2 × 10 mL) of the aqueous phase, dried (K2CO3)
and carefully concentrated at 10 Torr to give a yellow oil (85 mg). 1H
NMR with an internal standard (hexamethylbenzene, 10 mg) revealed
that the crude product contained a 4:1 ratio of 24 and 25 in 46% com-
bined yield. Column chromatography (hexane/Et2O, 96:4) allowed
the isolation of the major product, 24. The isolated yield of 24 was
only 10 mg (8%) due to the volatility of the compound.
26 (data are in good agreement with that reported18):
IR (film): ν = 2962 m, 2931 m, 2783 m, 1704 s, 1625 m, 1462 m,
1368 m cm–1.
UV (EtOH): λmax = 226 nm (log ε = 3.98).
1H NMR (270 MHz, CDCl3): δ = 5.86 (1 H, s, H-5), 3.23 (1 H, dd, J
= 11, 6 Hz, H1-eq), 2.97 (1 H, m, H-4), 2.80 (1 H, d + fs, J = 11 Hz,
H-3eq), 2.78 (1 H, obscured, H-7a), 2.30 (3 H, s, N-CH3), 2.18 (1 H,
dd, J = 11, 4 Hz, H-3ax), 1.94 (1 H, qd, J = 7, 3 Hz, H-7), 1.70 (1 H,
t, J = 11 Hz, H-1ax), 1.35 (3 H, d, J = 7 Hz, 4-CH3), 1.19 (3 H, d, J =
8 Hz, 7-CH3).
13C NMR (67.5 MHz, CDCl3): δ = 183.39 (0), 125.65 (1), 62.70 (2),
62.07 (2), 46.30 (1) and (3), 45.28 (1), 34.51 (1), 19.09 (3), 14.73 (3),
the carbonyl carbon could not be seen above the background noise.
CI-MS (NH3): m/z = 180 (MH+).
HRMS (CI, NH3): m/z found 180.1391 (MH+), C11H18NO requires
180.1388.
IR (film): ν = 3093 w, 2961 s, 2935 s, 2878 s, 2775 s, 1644 m, 1463
s, 1404 w, 1373 m, 891 s cm–1.
1H NMR (270 MHz, CDCl3): δ = 4.72 (1 H, s), 4.71 (1 H, s), 3.00 (1
H, dd, J = 11, 4 Hz), 2.89 (1 H, dd, J = 11, 4 Hz), 2.33 (1 H, m), 2.28
(3 H, s), 2.09 (1 H, m), 1.72 (1 H, ddq, J = 11, 4, 7 Hz), 1.61 (1 H, t,
J = 11 Hz), 1.55 (1 H, t, J = 11 Hz), 1.26 (1 H, d pentet, J = 14, 7 Hz),
1.03 (3 H, d, J = 7 Hz), 0.97 (3 H, t, J = 7 Hz).
(3S*,5S*)-3-Ethyl-1,5-dimethyl-4-[(E)-trimethylstannylmethyl-
ene]piperidine (31):
BuLi (2.5 M solution in hexanes, 1.0 mL, 2.5 mmol) was added to 3
(300 mg, 1.99 mmol) in THF (4 mL) at –40°C under argon and stirred
for 15 min. Me3SnCl (530 mg, 2.7 mmol) in THF (2 mL) was added
and the solution warmed to r.t. overnight. Sat. NaHCO3 solution
(10 mL) and Et2O (20 mL) were added and the mixture worked up as
normal to afford crude N-[(E)- and (Z)-but-2-enyl]-N-methyl-N-(2-
methyl-4-trimethylstannylbut-3-ynyl)amine (29) as a brown oil
(643 mg) which decomposed on attempted purification.
13C NMR (67.5 MHz, CDCl3): δ = 154.32 (0), 102.64 (2), 65.61 (2),
62.97 (2), 45.96 (3), 44.14 (1), 37.33 (1), 22.73 (2), 15.65 (3), 12.18
(3).
EI-MS: m/z (%) = 153 (M+, 69), 152 (71), 138 (35), 124 (47), 58
(100), 44 (92), 42 (58).
HRMS (EI): m/z found 153.1524 (M+), C10H19N requires 153.1517.

SYNTHESIS
Papers
556
To Cp2ZrCl2 (58 mg, 0.2 mmol) in THF (2 mL) at –78°C was added
BuLi (0.16 mL of 2.5 M soln in hexanes, 0.4 mmol) to give a yellow
solution. After 15 min crude 29 (60 mg) in THF (1 mL) was added and
the mixture allowed to warm to r.t. over 1 h. After stirring at r.t. for 2
h, MeOH (1mL) was added to the red-brown solution and the mixture
c) Kametani, T.; Suzuki, Y.; Ban, C.; Honda, T. Heterocycles
1987, 26, 1491.
(7) a) Kemp, M. I.; Whitby, R. J.; Coote, S. J. Synlett 1994, 451.
b) Kemp, M. I.; Whitby, R. J.; Coote, S. J. Synthesis, previous
paper in this issue.
poured into aq sat. NaHCO solution. Extractive workup (Et2O) and
(8) For initial communication of this work, see: Kemp, M. I.; Whit-
by, R. J.; Coote, S. J. Electronic Conference on Trends in Orga-
nic Chemistry (ECTOC-1) ISBN 0 85404 899 5; Rzepa, H. S.;
Goodman, J. M., Eds.; (CD-ROM), Royal Society of Chemistry
publications, 1995. See also http://www.ch.ic.ac.uk/ectoc/
(9) Negishi, E.; Holmes, S.J.; Tour, J.M.; Miller, J.A.; Cederbaum,
F.E.; Swanson, D.R.; Takahashi, T. J. Am. Chem. Soc. 1989,
111, 3336.
(10) For other total syntheses incorporating a zirconocene-induced
cocyclisation of an enyne followed by carbonylation, see:
a) Wender, P. A.; McDonald, F. E. Tetrahedron Lett. 1990, 31,
1543.
3
column chromatography (grade III basic alumina, petrol/Et2O, 97:3)
gave the title piperidine 31 (29 mg, 0.093 mmol, 50% from 3) as a
clear colourless oil.
IR (film): ν = 2964 s, 2932 s, 2877 m, 2839 m, 2777 m, 1600 m, 1462
m, 1445 m, 1376 m, 1310 m, 766 s cm–1.
1H NMR (270 MHz, CDCl3): δ = 5.40 (1 H, s), 2.98 (1 H, ddd, J = 10,
4, 2 Hz, H-2eq), 2.74 (1 H, d + fs, J = 11 Hz, H-6eq), 2.3–2.5 (2 H, m,
H-3 and H-5), 2.24 (3 H, s, NCH3), 2.06 (1 H, dd, J = 11, 4 Hz, H-
6ax), 1.69 (1 H, m, CH Me), 1.46 (1 H, t, J = 11 Hz, H-2ax), 1.24 (3
2
H, d, J = 7 Hz, 5-CH3), 1.19 (1 H, obscured, CH Me), 0.94 (3 H, t, J
2
= 7 Hz, CHCH3), 0.15 (9 H, s).
13C NMR (75 MHz, CDCl3): δ = 162.59 (0), 117.62 (1), 63.94 (2),
63.52 (2), 46.62 (3), 42.18 (1), 41.22 (1), 22.78 (2), 20.03 (3), 12.09
(3), –8.40 (3).
b) Agnel, G.; Owczarczyk, Z.; Negishi, E. Tetrahedron Lett.
1992, 33, 1543.
c) Negishi, E.; Ma, S. M.; Sugihara, T.; Noda, Y. J. J. Org.
Chem. 1997, 62, 1922.
d) Uesaka, N.; Saitoh, F.; Mori, M.; Shibasaki, M.; Okamura,
K.; Date, T. J. Org. Chem. 1994, 59, 5633.
CI-MS (NH3): m/z = 318 (MH+).
HRMS (CI, NH3): m/z found 318.1266 (MH+), C13H28NSn requires
318.1244.
(11) Pagenkopf, B. L.; Lund, E. C.; Livinghouse, T. Tetrahedron
1995, 51, 4421.
(12) Takahashi, T.; Kageyama, M.; Denisov, V.; Hara, R.; Negishi,
E. Tetrahedron Lett. 1993, 34, 687.
(13) Miura, K.; Funatsu, M.; Saito, H.; Ito, H.; Hosomi, A. Tetrahe-
dron Lett. 1996, 37, 9059.
A studentship from Glaxo Group Research to MIK is gratefully
acknowledged. We also thank Dr. John Langley for mass spectrome-
try, and Mrs. Joan Street for NMR studies.
(14) Urabe, H.; Hata, T.; Sato, F. Tetrahedron Lett. 1995, 36, 4261.
(15) N-Methylcrotylamine: Nordlander, J. E.; Catalene, D. B.; Eber-
lein, T. H.; Farkas, L. V.; Howe, R. S.; Stevens, R. M.; Tripou-
las, N. A.; Stansfield, R. E.; Cox, J. L.; Payne, M. J.; Viehbeck,
A. Tetrahedron Lett. 1978, 19, 4987.
(1) Hammouda, Y.; Motawi, M. M. Egypt. Pharm. Bull. 1959, 41,
73; Chem. Abstr. 1960, 54, 21646.
(2) Jones, G.; Fales, H. M.; Wildman, W. C. Tetrahedron Lett.
1963, 6, 397.
(16) Brandsma, L. Preparative Acetylenic Chemistry, Elsevier: Am-
sterdam, 1988; 2nd ed.
(3) Jones, G.; Ferguson, G.; Marsh, W. C. J. Chem. Soc., Chem.
Commun. 1971, 994.
(17) Gerard, F.; Miginiac, Ph. Synth. Commun. 1976, 6, 461.
(18) Alazard, J. P.; Leboff, A.; Thal, C. Tetrahedron Lett. 1989, 30,
5267.
(19) Manriquez, J. M.; McAlister, D. R.; Sanner, R. D.; Bercaw, J.
E. J. Am. Chem. Soc. 1978, 100, 2716.
(4) Hammouda, Y.; Rashid, A. K.; Amer, M. S. J. Pharm. Pharma-
col. 1964, 16, 834.
(5) Colin, G. G. J. Amer. Pharm. Ass. 1926, 15, 556.
(6) a) Imanishi, T.; Yagi, N.; Hanaoka, M. Chem. Pharm. Bull.
1983, 31, 1243; Tetrahedron Lett. 1981, 22, 667.
b) Miyashita, M.; Tanaka, D.; Shiratani, T.; Irie, H. Chem.
Pharm. Bull. 1992, 40, 1614.
(20) Castro, B.; Selve, C. Bull. Soc. Chim. Fr. 1971, 4368.